 Hello all, in this video we are going to see about p-value. Before calculating p-value for your research, you should understand about p-value clearly. For understanding p-value, firstly we should be clear about our research objective. If your study is about just estimating means or proportion in a specific population, you need not rather cannot calculate p-value. If you are studying relationship between two variables or comparing two groups, then you postulate and test the hypothesis. In any research we have two hypothesis, one is research hypothesis and null hypothesis. Research hypothesis or alternate hypothesis says there is a relationship between two variables or two groups, whereas null hypothesis says there is no relationship or null relationship between two variables or two groups. After framing this hypothesis, the next step is to apply the appropriate statistical test, which is based on the type of variable, type of distribution, sample size and unpaired or paired nature of the sample. So, after applying the statistical test, we get the test value of the statistical test. When this test value is approximated with the p-value tables, we can get the p-value or we can directly use statistical software which will give test value and p-value in its result. We are just one step away in understanding what is p-value. Here we are going to understand the type of errors. The null hypothesis may be actually true or false. In your study, you may accept that null hypothesis or reject that null hypothesis, which will derive four situations. When the null hypothesis is false and you reject it correctly, then it is called as true negative. When the null hypothesis is true and you accept it, that is called as true positive. When the null hypothesis is true and you reject it, that is called as false positive. When the null hypothesis is false and you accept it, that is called as false negative. So, this false positive is called as alpha error or type 1 error. negative is called as beta error or type 2 error. Alpha error is rejecting null hypothesis when it is actually true. Beta error is accepting null hypothesis when it is actually false. Accepting false is least dangerous when compared to rejecting the truth. So we always keep beta error at 10 percentage or 20 percentage before the start of the study and alpha level at 5 percentage or 1 percentage. The inverse of this beta error is called as power of the study. The probability of committing alpha error is called as p-value. As the probability of committing an alpha error is close to zero or the p-value is close to zero, the chance of you rejecting the null hypothesis when it is actually true is going to be nil. But conventionally people have kept it as 0.05 when the p-value is less than 0.05 you call it as statistically significant. When the p-value is less than 0.01 then you call it as highly significant. As equal to the statistical significance, clinical significance is also important. For example, when I mention certain variables like diastolic blood pressure, maximum decibels a person can hear, temperature of the body, all these values revolve around 9200. But when we calculate p-value for these values the statistical test remains same. But the clinical significance differs. So to sum up what is p-value? p-value is a measure of relationship or measure of association between two variables or groups. p-value is the probability of committing alpha error. p-value is the strength of evidence provided by our sample against null hypothesis. So p-value is the strength of evidence provided by our sample against null hypothesis in order to accept our alternate hypothesis. Thanks for watching this video. Please subscribe to think lateral channel. Thank you.